|Publication number||US5866809 A|
|Application number||US 08/784,024|
|Publication date||Feb 2, 1999|
|Filing date||Jan 15, 1997|
|Priority date||Feb 7, 1996|
|Also published as||DE69721096D1, DE69721096T2, EP0787619A2, EP0787619A3, EP0787619B1|
|Publication number||08784024, 784024, US 5866809 A, US 5866809A, US-A-5866809, US5866809 A, US5866809A|
|Original Assignee||Scania Cv Aktiebolag|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (2), Referenced by (12), Classifications (18), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
The invention relates to a method for correction of engine torque during automated gear changes of mechanical stepped gearboxes in order to reduce the torque at gear contact points.
Automatic gear changes of mechanical stepped gearboxes require adjustment of the torque delivered from the engine in order to reduce the torque transferred at the point of contact of the relevant gears. At the time of gear disengagement it is desirable for there to be a torque-free state at the contact point between the relevant gears in the gearbox, but since it is not practically possible to measure the torque at the gear contact point the torque adjustment has to be achieved indirectly by adjusting the torque delivered from the engine.
SE-A-9401653-9 which corresponds to U.S. Pat. No. 5,595,551 refers to a solution whereby gear changing takes place without the disc clutch being released. In this case a gear change is preceded by the engine torque being adjusted towards a zero-torque level with the intention of keeping the engine speed constant during the gear change. The torque with which the engine is modulated is calculated on the basis of available data concerning the engine, its moment of inertia and its internal friction and the like. The modulated torque may also depend on whether any power take-off is operating or not.
A disadvantage of this method is that there is no direct measurement of the torque and that the modulation is based on the expected behavior of the engine. Whether the modulated torque is correct or not is not known. The manufacture of engines and various driveline components may involve tolerance differences which result in each individual vehicle having different characteristics. Even if the behavior of an engine can be predicted correctly when it is new, its characteristics will change over time and with engine wear. This means that the engine torque modulated over time is probably not the same as when the engine was new. This in its turn means that gear changing may become difficult and/or that it is unacceptably delayed in certain circumstances. Oscillations in the control system may result in a very long scaling-in time being required for being able to determine correctly the fuel quantity which gives constant synchronous speed.
The object of the invention is to be able easily and quickly, on the basis of wear in the system and disparities between different individual systems, to correct a predetermined engine torque during gear changes without using any power sensor or torque sensor, so that there is a torque-free state in the gearbox during gear disengagement. To this end, the invention concerns a process for correcting the torque to which an engine is adjusted before a gear disengagement, preferably on the occasion of automated gear change of a stepped gearbox without using any clutch. Immediately after a gear has been disengaged, the engine is adjusted towards a zero torque (M0) which will maintain the engine speed (n) constant during the gear disengagement. This zero-torque level is stored in a matrix for the relevant operating point given by at least the engine speed (n) and the engine temperature (T). During a fixed measuring period after the gear disengagement when the gearbox is in neutral position, the acceleration of the engine when it is run with the relevant zero torque is detected. Thereafter any acceleration detected is converted to a corresponding torque, which represents an error in the modulated zero torque. The original zero-torque value is thereafter corrected by applying the detected incorrect torque so that before a subsequent gear disengagement the engine is modulated with a more correct value of the zero torque (M0).
Other features distinguishing the invention are indicated in the description below of an embodiment with reference to the attached drawings.
FIG. 1 is a skeleton diagram of the components of a gear change system for mechanical stepped gearboxes.
FIG. 2 is a flow diagram for engine torque adjustment during disengagement of a gear in accordance with the invention.
FIGS. 3, 4 and 5 respectively show how engine speed and engine torque vary during a gear change when the zero torque modulated during gear disengagement is respectively too high, too low or correct.
FIG. 1 depicts a gear change system for monitoring and controlling computer-assisted gear changing of mechanical gearboxes in a motor vehicle. The composition and operation of the system correspond largely tothe system described in the previously mentioned SE,A,9401653-2, which corresponds to U.S. Pat. No. 5,595,551, incorporated herein by reference so the only items depicted are those required for understanding the present invention.
The vehicle is powered by a combustion engine 2, preferably a diesel engine, which is connected to the vehicle's driving wheels 10 via a clutch4, a mechanical stepped gearbox 6 and a propeller shaft 8. In this embodiment the clutch 4 is only intended to be operated manually at low vehicle speeds and when starting and stopping. In gear changing during operation of the vehicle between different running gears, the clutch 4 is intended to be operated neither manually nor automatically, so gear changes take place with the clutch 4 acting as a drive power transmission connection between the engine and the gearbox 6.
The gear change system effects gear changes partly by adjusting the engine speed and engine torque at gear changes and partly by operating the servosin the gearbox which disengage the operative gear and engage the next gear.
The clutch 4 not being released during gear changes results in severe engine control system requirements if gear changes are to take place with the shortest possible torque break in the mechanical stepped gearbox both when what is required is the possibility of torque-free disengagement of agear and when engaging the next gear, since the engine control system has to ensure that the synchronous speed for the next gear is reached before the gear is engaged.
The gear change system incorporates a control unit 12 with microcomputer which is connected by different lines to various parts of the system. These connections are used for transmitting various signals corresponding to what is illustrated by arrows in FIG. 1. The control unit 12 also has two-way connections in that it sends output signals and receives input signals from a multiplicity of control units via a number of links represented in FIG. 1 by bi-directional arrows.
Via a link 14 with the engine fuel injection system 15, which is in its turn connected to various sensors, the control unit 12 receives information on the engine speed n, from which the engine acceleration a may also be calculated. The control unit 12 also receives via the link 14 information on the modulated engine torque M, which is in practice calculated by means of the fuel quantity injected. The fuel injection system 15 in its turn controls via the link 16 the fuel quantity to the engine injectors. During gear changes the control unit 12 provides the fuel injection system 15 with signals which cause the engine to be modulated with a certain torque M.
The control unit 12 receives from a temperature sensor via a line 17 a signal representing the engine temperature T or, in practice, the temperature of the engine coolant. A sensor 18 provides the control unit 12 via a signal line 19 with information on whether any power take-off (PTO) is operating. In practice this sensor 18 may consist of the circuit-breaker via which the power take-off is connected. In cases where a number of power take-offs are arranged to be driven, they may be correspondingly connected to the control unit, but in this embodiment it is supposed that only one power take-off is involved.
The control unit 12 is connected via a link 7 to various solenoid valves inthe gearbox 6 which activate the servos for engagement and disengagement ofgears. This link 7 is also used for providing the control unit 12 with signals representing the operating state of the gearbox at the time and indicating which gear is engaged, and, during gear changes, with signals representing the various phases of the gear change.
The control unit 12 is also connected to other vehicle controls (not shown), e.g., gear selector, accelerator pedal, brake pedal, retarder. As the latter operate conventionally and make no difference to the process according to the invention, no further description of them is provided.
Gear changes are initiated entirely automatically if the driver selects an automatic position or manually if the driver selects a manual position. Regardless of how gear changing is initiated (automatically by the controlsystem or manually by the driver), it is effected by the control unit 12 without the clutch 4 having to be released.
As stated in the introduction, it is important when disengaging gears to ensure that the contact point of the relevant gears is not loaded with anytorque. The adjustment of engine torque to the so-called zero-torque level before the disengagement of a gear is described in more detail with reference to the flow diagram in FIG. 2. The latter also shows the processaccording to the invention whereby gear disengagement is followed by correction of the factor MG which determines the zero-torque level orof the power takeoff factor MPTO which affects the zero-torque level. This torque correction routine is stored in the microcomputer of the control unit 12.
Gear changing is initiated at step 20, after which the control unit 12 starts adjusting the engine torque down to the zero torque M0. The zero torque depends on the engine speed n and the engine temperature T. The control unit therefore has stored in it a basic matrix with various values of the zero torque MG pertaining to various engine speeds n and temperatures T. At the time of manufacture of the vehicle this basic matrix is stored in the control unit 12 with empirically established values. To compensate for the increased torque requirement during operation of the power take-off, this basic value MG has to be supplemented by a factor pertaining to the running of the power take-off. This power take-off factor MPTO likewise depends on the engine speed n and the engine temperature T, so the control unit 12 has stored in it a corresponding matrix for various MPTO factors.
The opening step 21 decides whether the power take-off is operating or not.If no power take-off is operating, the adjustment of the zero torque takes place according to the lefthand part of FIG. 2, whereas if one or more power take-offs are operating the adjustment takes place according to the righthand part of the diagram. If no power take-offs are operating, step 22 adjusts the engine torque M to the zero-torque value MG stored in the basic matrix.
When step 23 finds that the torque has become equal to that desired, step 24 disengages the gear, which is then supposed to take place with zero torque at the gear contact point.
According to the present invention, gear disengagement is directly followedby measurement of the change in the engine speed n during a limited measuring period mp. This measuring period covers a time when the gearbox is in neutral position before the engine is modulated to a suitable speed for the engagement of the next gear. The measuring period thus covers a period of time when the engine is modulated at what is supposed to be a correct zero torque MG. What takes place during the measuring period and directly after it is shown in a common step 25. The measurement of theengine speed change Δn over the measuring period can be combined withknowledge of the measuring period duration Δt to calculate the engine's corresponding acceleration am. This in its turn means that on the basis of the engine's moment of inertia J it is possible to calculate, by the formula MERR =am *J, the torque which causes the speed change. This torque corresponds to the amount by which the supposed torque MG was incorrect. The value of MERR will correspond in amount to the value by which the modulated torque has to be corrected. Step 26 thereafter effects an updating of the MG value stored in the basic matrix by subtracting from the previously stored MG value the incorrect value MERR which is stored as a new valueof MG and which in a subsequent adjustment provides a better and more correct value of MG.
As these torques apply for a certain engine speed n and a certain engine temperature T, the updating is only effected for the relevant speed and temperature values. The process according to the invention for updating the matrix containing the zero-torque values is then complete, as illustrated in FIG. 2 by a final step 27.
If step 21 finds that the power take-off is operating, a higher zero torqueMG is required. Step 28 then modulates the engine towards a zero torque which is the aggregate of a basic torque MG derived from the basic matrix and a torque MPTO derived from the power takeoff matrix.When this zero torque is reached at step 29, step 30 effects disengagement of the gear. Thereafter step 31 carries out, in a manner analogous with step 25, a calculation of the incorrect torque MERR. This is followedby step 32 updating the power take-off matrix containing the MPTO torque values. It may be noted that the updating takes place in the lattermatrix and not in the basic matrix containing MG torque values. This is because the number of gear changes with no power take-off activated is likely to be significantly greater and because it may be assumed that the basic matrix is probably more correct. The MG value adopted at step 28 may therefore be supposed to be correct, so the engine speed changes measured during the measuring period are primarily due to incorrect valuesin the power take-off matrix.
FIGS. 3 to 5 illustrate the gear change pattern described and show how the engine speed may vary therein. The time notations used are the same in allthree diagrams. The gear change initiated at time t0 results in the torque M being modulated towards the relevant zero-torque value, either MG or MG +Mp TO. At time t1 the relevant zero torque is reached and the gear is disengaged. Up to this point the engine speed nhas been constant because there has been no change in the vehicle speed. The gear disengagement at time t1 marks the start of the measuring period mp, which in this case is the same length of time as the period of engine modulation by the zero torque and runs to time t2. As may be seen in FIG. 4, the engine speed has increased over the measuring period by Δn, which means that the modulated zero torque was too great, whereas FIG. 5 shows the engine speed having decreased by Δn, which means that the modulated torque was too little. In contrast, FIG. 3 shows the engine speed n remaining constant over the measuring period, so in this case the modulated torque was correct. At time t2 the torque M is ramped down further with a view to reducing the engine speed n to the correct value before the next gear is engaged. In this case the engine speed will decrease, since there is supposed to be an upshift to a higher gear whereby the engine speed after the gear change will be lower than theengine speed before the gear change. Had a downshift been involved, the torque would instead have increased at time t2 in order to increase the engine speed n before the gear engagement. At time t3 the correctengine speed for the next gear has been reached and the gear engagement starts, being completed at time t4. At time t4 the torque is increased so that at time t5 it corresponds to the torque which the driver requires. The whole gear change pattern from time t0 to time t5 is controlled by the control unit 12, and the driver's operation of various controls during that period has no effect on the gear change pattern. After time t5 both torque and engine speed revert to being controlled by an accelerator control operable conventionally by the driver.
The process described for correcting the zero torque can take place continuously during each gear change, but since over time ever fewer and smaller corrections will be required, it is advantageous after a certain time to reduce the number of corrections so that they are not effected at every gear change. The control unit therefore has stored in it a log matrix which indicates how many updates have been effected at the various operating points in the two matrices. When the system is new it is important that it should adapt quickly to the particular engine and, whereapplicable, the particular power take-off(s) . Corrections are therefore always initiated so long as the matrix for the particular operating point does not indicate that a predetermined number of corrections have already been effected. This predetermined number of corrections is determined by how great the deviations are between individual systems, but its order of magnitude in the case of mass-produced systems may be taken to range from a few tens of corrections to 100 corrections.
When the individual system has been subjected to suitable adjustment, the need for initiation of corrections will be less frequent and they will only adjust the torque according to subsequent wear etc. The log matrix may, for example, be used thereafter for counting all gear changes which take place at each operating point, and the initiation of corrections may subsequently take place on a longer time basis depending on the value in the log matrix. For example, correction may be initiated at every gear change for the first 50 gear changes, at every tenth gear change for gear changes 51-300, at every twentieth gear change for gear changes 301-1000, and so on.
Experiments have shown that the length of the preferably fixed measuring period mp can be kept as short as 0.1 second in certain applications. In practical use, however, there may be good reasons for adopting somewhat longer measuring periods, e.g., between 0.1 and 0.4 second, preferably 0.3second, because the system may be subject to signal delays and inertia in mechanical components.
The MG (n,T) and MPTO (n,T) matrices contain predetermined operating points whereby the operating points with regard to engine temperature T are divided into steps of 10°-20° C. in the normal engine operating temperature range of about 70°-90° C. and into smaller steps at lower operating temperatures in the range below 70° C., and those with regard to engine speed are divided into steps of 100-200 rpm.
In the correction of the MG (n,T) and MPTO (n,T) matrices at eachrelevant operating point at which gear changing and measurement are initiated it is also possible to correct nearby ones (e.g., at least four nearby operating points in the particular matrix) proportionally to the relevant correction torque and to how close the nearby operating points inthe matrix are to the particular operating point. If the particular operating point corresponds directly to the operating point in the particular matrix, only the latter operating point is corrected.
In the embodiment described, correction of the zero torques MG and MPTO was effected in the respective matrices so that in each matrix anew corrected value replaced a previous stored value. In an alternative embodiment it is in each case possible instead to have fixed basic values stored in a basic matrix and have the required corrections stored in a separate correction matrix so that the relevant torque value is formed by the values from these two matrices being combined with one another. In such cases all that is updated is the correction required in the correction matrix.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5456643 *||Jun 10, 1993||Oct 10, 1995||Honda Giken Kogyo Kabushiki Kaisha||System for controlling automobile transmission|
|US5595551 *||May 8, 1995||Jan 21, 1997||Scania Cv Aktiebolag||Method for control of engine torque during gear changing|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US6035252 *||Sep 29, 1998||Mar 7, 2000||Ford Global Technologies, Inc.||Engine torque control|
|US6155955 *||Nov 3, 1997||Dec 5, 2000||Zf Friedrichshafen Ag||Operating method for a motor vehicle driving unit|
|US6345529 *||Jun 24, 1998||Feb 12, 2002||Scania Cv Aktiebolag||Method of gear changing in a motor vehicle|
|US7507181||May 15, 2008||Mar 24, 2009||Bayerische Motoren Werke Aktiengesellschaft||Method for determining a driving torque correction factor for compensating cooperating driving torques of different drive devices|
|US7769514 *||Feb 25, 2005||Aug 3, 2010||Renault S.A.S.||Method and device for damping control of oscillating modes of a continuously variable transmission provided with an electric variator|
|US7860630 *||Dec 8, 2005||Dec 28, 2010||Renault S.A.S.||Method and device for controlling heat engine torque saturation in an infinitely variable transmission with electric speed selector|
|US8095285||Jun 29, 2007||Jan 10, 2012||Caterpillar Inc.||Method for derating a power source to limit damage|
|US20050159270 *||Mar 17, 2003||Jul 21, 2005||Lars-Gunnar Hedstrom||Arrangement and method for allowing disengagement of a gear in a gearbox|
|US20060047395 *||Nov 7, 2003||Mar 2, 2006||Mitsubishi Fuso Truck And Bus Corporation||Method and device for controlling gear shift of mechanical transmission|
|US20070192008 *||Feb 25, 2005||Aug 16, 2007||Renault S.A.S||Method and device for damping control of oscillating modes of a continuously variable transmission provided with an electric variator|
|US20080146414 *||Dec 8, 2005||Jun 19, 2008||Renault S.A.S.||Method And Device For Controlling Heat Engine Torque Saturation In An Infinitely Variable Transmission With Electric Speed Selector|
|US20090000292 *||Jun 29, 2007||Jan 1, 2009||Schifferer Andrew N||Method for derating a power source to limit damage|
|U.S. Classification||73/114.15, 73/862.31, 73/862.191|
|International Classification||B60W30/18, B60W10/11, F16H61/04, F02D29/00, B60W10/04, F16H61/00, F16H63/50|
|Cooperative Classification||F16H2342/00, B60W10/11, F16H63/502, B60W10/04, B60W2510/1065, B60W30/1819, B60W2510/0676|
|Jan 15, 1997||AS||Assignment|
Owner name: SCANIA CV AKTIEBOLAG, A SWEDISH CORPORATION, SWEDE
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:SODERMAN, GORAN;REEL/FRAME:008397/0985
Effective date: 19970115
|Jul 11, 2002||FPAY||Fee payment|
Year of fee payment: 4
|Jul 7, 2006||FPAY||Fee payment|
Year of fee payment: 8
|Jul 1, 2010||FPAY||Fee payment|
Year of fee payment: 12